Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 142 of the 142 references it cites were found in a registry or an index.

Co-design of Human-centered, Explainable AI for Clinical Decision Support

Open access
Panigutti, Cecilia, Beretta, Andrea, Fadda, Daniele, Giannotti, Fosca, Pedreschi, Dino, Perotti, Alan, Rinzivillo, Salvatore · 2023 · ACM Transactions on Interactive Intelligent Systems

Abstract

eXplainable AI (XAI) involves two intertwined but separate challenges: the development of techniques to extract explanations from black-box AI models and the way such explanations are presented to users, i.e., the explanation user interface. Despite its importance, the second aspect has received limited attention so far in the literature. Effective AI explanation interfaces are fundamental for allowing human decision-makers to take advantage and oversee high-risk AI systems effectively. Following an iterative design approach, we present the first cycle of prototyping-testing-redesigning of an explainable AI technique and its explanation user interface for clinical Decision Support Systems (DSS). We first present an XAI technique that meets the technical requirements of the healthcare domain: sequential, ontology-linked patient data, and multi-label classification tasks. We demonstrate its applicability to explain a clinical DSS, and we design a first prototype of an explanation user interface. Next, we test such a prototype with healthcare providers and collect their feedback with a two-fold outcome: First, we obtain evidence that explanations increase users’ trust in the XAI system, and second, we obtain useful insights on the perceived deficiencies of their interaction with the system, so we can re-design a better, more human-centered explanation interface.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1145/3587271

Cites

142 of 142 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 (1st plot) A representation of a branch of the tree-shaped ICD-9 hierarchical ontology: The root is a general condition Disease, while its children and grandchildren are increasingly more specific conditions. (2nd plot) . A simple example of the representation a patient clinical history as a sequence of ICD-9 codes.
Fig. 1 (1st plot) A representation of a branch of the tree-shaped ICD-9 hierarchical ontology: The root is a general condition Disease, while its children and grandchildren are increasingly more specific conditions. (2nd plot) . A simple example of the representation a patient clinical history as a sequence of ICD-9 codes.CC-BY · From the PDF
Fig. 2 The explanation pipeline in the case of ontology-linked data (blue) and not ontology-linked data (red).
Fig. 2 The explanation pipeline in the case of ontology-linked data (blue) and not ontology-linked data (red).CC-BY · From the PDF
Fig. 3 (1st plot) The node corresponding to the randomly selected ICD-9 code ( 276 . 4 ) of the patient is highlighted in red in the ICD-9 ontology graph representation. (2nd plot) . The ontological superconcept of the selected ICD-9 is selected and highlighted ( 276 ). (3rd plot) All ICD-9 codes all having as parent the identified superconcept are selected and removed from the patient (codes 276 . 1 , 276 . 2 , and 276 . 4 ).
Fig. 3 (1st plot) The node corresponding to the randomly selected ICD-9 code ( 276 . 4 ) of the patient is highlighted in red in the ICD-9 ontology graph representation. (2nd plot) . The ontological superconcept of the selected ICD-9 is selected and highlighted ( 276 ). (3rd plot) All ICD-9 codes all having as parent the identified superconcept are selected and removed from the patient (codes 276 . 1 , 276 . 2 , and 276 . 4 ).CC-BY · From the PDF
Fig. 6 Fidelity distribution at k = 10 , 20 , 30 for the explanation of the multi-label classification task. Each color represents a different condition.
Fig. 6 Fidelity distribution at k = 10 , 20 , 30 for the explanation of the multi-label classification task. Each color represents a different condition.CC-BY · From the PDF
Fig. 7 Fidelity distribution for the explanation of the binary classification task. Each color represents a different condition. (1st plot) results for the CCS code 100, acute myocardial infarction . (2nd plot) results for the CCS code 103, pulmonary heart disease .
Fig. 7 Fidelity distribution for the explanation of the binary classification task. Each color represents a different condition. (1st plot) results for the CCS code 100, acute myocardial infarction . (2nd plot) results for the CCS code 103, pulmonary heart disease .CC-BY · From the PDF
Fig. 8 Doctor XAI interface for binary prediction tasks.
Fig. 8 Doctor XAI interface for binary prediction tasks.CC-BY · From the PDF
Fig. 9 Flowchart of the estimation task for the two interfaces: only suggestion (blue path) and suggestion and explanation (yellow path).
Fig. 9 Flowchart of the estimation task for the two interfaces: only suggestion (blue path) and suggestion and explanation (yellow path).CC-BY · From the PDF
Fig. 10 Boxplot comparing the WOA (a), the confidence shift after the advice (b), the behavioral intention of use (c), and the explicit trust in the two systems (d).
Fig. 10 Boxplot comparing the WOA (a), the confidence shift after the advice (b), the behavioral intention of use (c), and the explicit trust in the two systems (d).CC-BY · From the PDF
Fig. 11 New explanation user interface.
Fig. 11 New explanation user interface.CC-BY · From the PDF

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 119 references

Research works printed but not linked: 8

  1. Barbara D. Adams, Lora E. Bruyn, Sébastien Houde, Paul Angelopoulos, Kim Iwasa-Madge, and Carol McCann. 2003. Trust in automated systems. Minist. Nat. Defen. (2003).
  2. Tal Baumel, Jumana Nassour-Kassis, Raphael Cohen, Michael Elhadad, and Noémie Elhadad. 2018. Multi-label classification of patient notes: Case study on ICD code assignment. In Proceedings of the Workshops at the 32nd AAAI Conference on Artificial Intelligence.
  3. Donald J. Berndt and James Clifford. 1994. Using dynamic time warping to find patterns in time series. In Proceedings of the 3rd International Conference on Knowledge Discovery and Data Mining (AAAIWS’94). AAAI Press, 359–370. Retrieved from http://dl.acm.org/citation.cfm?id=3000850.3000887.
  4. Edward Choi, Mohammad Taha Bahadori, Andy Schuetz, Walter F. Stewart, and Jimeng Sun. 2016. Doctor AI: Predicting clinical events via recurrent neural networks. In Proceedings of the Machine Learning for Healthcare Conference. PMLR, 301–318.
  5. Ian Covert, Scott M. Lundberg, and Su-In Lee. 2021. Explaining by removing: A unified framework for model explanation. J. Mach. Learn. Res. 22 (2021), 209–1.
  6. ISO 9241-210:2019. 2019. Ergonomics of Human-system Interaction Human-centred Design for Interactive Systems. Standard. International Organization for Standardization.
  7. Jennifer Preece, Yvonne Rogers, and Helen Sharp. 2019. Interaction Design: Beyond Human-Computer Interaction (5th ed.). Wiley.
  8. Sana Tonekaboni, Shalmali Joshi, Melissa D. McCradden, and Anna Goldenberg. 2019. What clinicians want: Contextualizing explainable machine learning for clinical end use. In Proceedings of the Machine Learning for Healthcare Conference. PMLR, 359–380.

Cited by

See all 159

Metadata

Authors
Panigutti, C., Beretta, A., Fadda, D., Giannotti, F., Pedreschi, D., Perotti, A., & Rinzivillo, S.
Affiliations
  • University of Pisa (IT)
  • National Research Council (IT)
  • Scuola Normale Superiore (IT)
  • CENTAI Institute, Italy
Venue
ACM Transactions on Interactive Intelligent Systems
Published
2023-12-08
Volume
13
Issue
4
Pages
1-35
ISSN
  • 2160-6455 (ISSN-L)
  • 2160-6463
OpenAlex
W4324155120
Cited by
127
References
119
Citations by year
2026: 56 · 2025: 37 · 2024: 29 · 2023: 5
FWCI
11.97
Type
journal article
Language
en
Open access
Locations
  • ACM Transactions on Interactive Intelligent Systems (publishedVersion, cc-by)
  • CINECA IRIS Institutial research information system (University of Pisa) (submittedVersion, other-oa)
  • ISTI Open Portal (submittedVersion, other-oa)
  • Scuola Normale Superiore di Pisa (submittedVersion, other-oa)
Funders
  • European Union (ERC-2018-ADG G.A. 834756 (XAI))
  • HumanE AI Net (952026)
  • PNRR - M4C2 - Investimento 1.3, Partenariato Esteso (PE00000013)
  • UK government (10061955)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
Keywords
explainable AI, clinical decision support systems, human-centered design, trust in AI, multi-label classification, iterative prototyping

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Co-design of Human-centered, Explainable AI for Clinical Decision Support. https://app.cochat.ai/papers/10.1145/3587271
The paper
Panigutti, C., Beretta, A., Fadda, D., Giannotti, F., Pedreschi, D., Perotti, A., & Rinzivillo, S. (2023). Co-design of Human-centered, Explainable AI for Clinical Decision Support. ACM Transactions on Interactive Intelligent Systems. https://doi.org/10.1145/3587271

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.